Many drugs administered to treat a disease are targeted against general differences between a diseased cell and a normal cell. For example, paclitaxel, which is used to treat ovarian and breast cancer and inhibits microtubule function, is thought to exhibit tumor cell specificity based on the greater rate of proliferation of tumor cells relative to normal cells (Miller and Ojima, Chem. Rec. 1:195-211, 2002). However, despite this consensus view, paclitaxel's in vitro activity varies widely across tumor cell lines (Weinstein et al., Science 275:343-349, 1997), indicating that genetic factors can modify sensitivity of tumor cells to paclitaxel and that the responsiveness of tumor cells is not simply determined by their rate of proliferation.
Molecularly targeted therapeutics represent a promising new approach to anti-cancer drug discovery (Shawver et al., Cancer Cell 1: 117-23, 2002). Using this approach, small molecules are designed to inhibit directly the very oncogenic proteins that are mutated or overexpressed in specific tumor cell types. By targeting specific molecular defects found within tumor cells, this approach may ultimately yield therapies tailored to each tumor's genetic makeup. Two recent examples of successful molecularly targeted anti-cancer therapeutics are Gleevec (imatinib mesylate), an inhibitor of the breakpoint cluster region-abelsen kinase (BCR-ABL) oncoprotein found in Philadelphia chromosome-positive chronic myelogenous leukemia (Capdeville et al., Nat Rev Drug Discov 1: 493-502, 2002) and Herceptin (trastuzumab), a monoclonal antibody targeted against the HER2/NEU oncoprotein found in metastatic breast cancers (Mokbel and Hassanally, Curr Med Res Opin 17: 51-9, 2001).
A complementary strategy involves searching for genotype-selective anti-tumor agents that become lethal to tumor cells only in the presence of specific oncoproteins or in the absence of specific tumor suppressors. Such genotype-selective compounds might target oncoproteins directly or they might target other critical proteins involved in oncoprotein-linked signaling networks. Compounds that have been reported to display synthetic lethality include (i) the rapamycin analog CCI-779 in myeloma cells lacking PNEN (Shi et al., Cancer Res 62: 5027-34, 2002), (ii) Gleevec in BCR-ABL-transformed cells (Druker et al., Nat Med 2: 561-6, 1996) and (iii) a variety of less well-characterized compounds (Stockwell et al., Chem Biol 6: 71-83, 1999; Torrance et al., Nat Biotechnol 19: 940-5, 2001).
Despite the research discussed above, there remains a significant need to develop and/or identify compounds that selectively target tumor cells.